Research map / Point clouds / Point cloud classification
Upright-Net: Learning Upright Orientation for 3D Point Cloud
Xufang Pang, Feng Li, Ning Ding, Xiaopin Zhong
Check your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Estimate the natural base by point segmentation, then derive orientation from its fitted plane | Upright-Net · UprightNet15 test | Accuracy (%): 69.01 -> 91.80 | Improved |
| Estimate the natural base by point segmentation, then derive orientation from its fitted plane | Upright-Net · UprightNet5 mean | Accuracy: 28.40 -> 81.39 | Improved |
| Use Upright-Net for classification on rotationally augmented data | RobustPointSet · Rot/Rot | Classification accuracy: 58.95 -> 67.83 | Improved |
| Pre-train the Upright-Net encoder before downstream classification | RobustPointSet · Org/Rot | Classification accuracy: 14.18 -> 49.92 | Improved |
| Pre-train the Upright-Net encoder before downstream classification | RobustPointSet · Rot/Rot | Classification accuracy: 67.83 -> 79.62 | Improved |
| Replace the proposed segmentation backbone with PointNet (ablation) | Upright-Net · UprightNet15 test | Accuracy (%): 91.80 -> 77.68 | Hurt |
| Replace the proposed segmentation backbone with DGCNN (ablation) | Upright-Net · UprightNet15 test | Accuracy (%): 91.80 -> 90.26 | Hurt |
| Add fitting-residual and stability loss terms rather than using BCE alone (ablation) | Upright-Net · UprightNet15 test | Accuracy (%): 85.13 -> 91.80 | Improved |
| Use α1=1 and α2=0.1 instead of the reported default α1=0.1 and α2=1 (ablation) | Upright-Net · UprightNet15 test | Accuracy (%): 91.80 -> 92.47 | Improved |
| Use α1=1 and α2=1 instead of the reported default α1=0.1 and α2=1 (ablation) | Upright-Net · UprightNet15 test | Accuracy (%): 91.80 -> 91.22 | Hurt |
Closest papers by meaning
- UprightNet: Geometry-Aware Camera Orientation Estimation From Single Images (ICCV 2019 · similarity 0.90)
- Rotation-Adaptive Point Cloud Domain Generalization via Intricate Orientation Learning (TPAMI 2025 · similarity 0.88)
- Deep Positional and Relational Feature Learning for Rotation-Invariant Point Cloud Analysis (ECCV 2020 · similarity 0.88)
- Symmetry-Robust 3D Orientation Estimation (ICML 2025 · similarity 0.87)
- Stable and Consistent Prediction of 3D Characteristic Orientation via Invariant Residual Learning (ICML 2023 · similarity 0.87)
- GAPrompt: Geometry-Aware Point Cloud Prompt for 3D Vision Model (ICML 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
